Your browser doesn't support javascript.
loading
Fine-grained diabetic wound depth and granulation tissue amount assessment using bilinear convolutional neural network.
Zhao, Xixuan; Liu, Ziyang; Agu, Emmanuel; Wagh, Ameya; Jain, Shubham; Lindsay, Clifford; Tulu, Bengisu; Strong, Diane; Kan, Jiangming.
Afiliación
  • Zhao X; School of Technology, Beijing Forestry University, Beijing, China, 100083.
  • Liu Z; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Agu E; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Wagh A; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Jain S; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Lindsay C; Radiology Department, University of Massachusetts Medical School, Worcester MA, USA, 01655.
  • Tulu B; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Strong D; Computer Science Department, Worcester Polytechnic Institute, Worcester, MA, USA, 01609.
  • Kan J; School of Technology, Beijing Forestry University, Beijing, China, 100083.
IEEE Access ; 7: 179151-179162, 2019.
Article en En | MEDLINE | ID: mdl-33777590
Diabetes mellitus is a serious chronic disease that affects millions of people worldwide. In patients with diabetes, ulcers occur frequently and heal slowly. Grading and staging of diabetic ulcers is the first step of effective treatment and wound depth and granulation tissue amount are two important indicators of wound healing progress. However, wound depths and granulation tissue amount of different severities can visually appear quite similar, making accurate machine learning classification challenging. In this paper, we innovatively adopted the fine-grained classification idea for diabetic wound grading by using a Bilinear CNN (Bi-CNN) architecture to deal with highly similar images of five grades. Wound area extraction, sharpening, resizing and augmentation were used to pre-process images before being input to the Bi-CNN. Innovative modifications of the generic Bi-CNN network architecture are explored to improve its performance. Our research generated a valuable wound dataset. In collaboration with wound experts from University of Massachusetts Medical School, we collected a diabetic wound dataset of 1639 images and annotated them with wound depth and granulation tissue grades as labels for classification. Deep learning experiments were conducted using holdout validation on this diabetic wound dataset. Comparisons with widely used CNN classification architectures demonstrated that our Bi-CNN fine-grained classification approach outperformed prior work for the task of grading diabetic wounds.
Palabras clave

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: IEEE Access Año: 2019 Tipo del documento: Article Pais de publicación: Estados Unidos

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: IEEE Access Año: 2019 Tipo del documento: Article Pais de publicación: Estados Unidos